The overall goal of this proposal is to identify the genetic changes responsible for initiation of ovarian cancer. The identification of the mutations that occur early in tumorigenesis could enable the development of molecular approaches for earlier diagnosis, at the time when current therapeutic strategies have a high cure rate. Since the overwhelming majority of ovarian tumors arise from the surface epithelium, the development of ovarian surface epithelial cell lines would serve as a strong foundation for the molecular analysis of events surrounding the initiation of ovarian carcinogenesis. To identify the genetic changes responsible for the earliest steps in malignant transformation in the absence of the multiple molecular and cytogenetic alterations commonly observed in advanced ovarian tumors, a cell culture model will be utilized (Specific Aim 1: Induce spontaneous transformation of ovarian surface epithelial cells from oophorectomy specimens obtained from individuals with different relative risks for development of ovarian cancer). Preliminary studies included in this proposal indicate that human ovarian surface epithelial cells isolated and placed in tissue culture to stimulate growth can undergo spontaneous malignant transformation. Furthermore, these cell lines show molecular and cytogenetic changes which are consistent with the genetic abnormalities observed in ovarian carcinomas, suggesting that this cell culture system has the potential to identify candidate genes involved in ovarian carcinogenesis. The work proposed, is designed to carefully evaluate the genetic changes observed in spontaneously transformed human surface epithelial cells by LOH analysis, whole genome scanning, and high resolution cytogenetic techniques and to determine their relationship to the abnormalities found in fresh ovarian tumors (Specific Aim 2: Compare the allelotypes of malignantly transformed human ovarian surface epithelial cells and ovarian tumors in order to differentiate the genetic changes associated with disease initiation from those which arise during progression and Specific Aim 3: Use whole genome scanning and cytogenetic approaches to identify the location of candidate gene(s) responsible for the initiation of the malignant phenotype of human ovarian surface epithelial cells). These approaches have the potential to identify critical genetic changes relevant to initiation of clinical ovarian cancer, all in the absence of the numerous genetic abnormalities characteristic of ovarian cancer at diagnosis. The long range goal is to determine the causal role in ovarian carcinogenesis of these candidate genes identified through implementation of this proposal. Therefore, these studies should provide important insights into the specific genetic alterations involved in the pathogenesis of ovarian cancer, increase the understanding of the basic biology of this disease, and ultimately benefit future patients.